Atherosclerosis and related complications are among the disorders contributing to the highest rates of human mortality and have become a substantial threat to human health(Reference Neeland, Ross and Després1). The size of aortic lesions, accompanied by elevated levels of serum cholesterol and aortic inflammation, has been found to be a key variable in atherosclerosis(Reference Markin, Sobenin and Grechko2). Numerous clinical studies have shown that reducing serum cholesterol and aortic endothelial inflammation are salient factors in the prevention and treatment of this disease(Reference Jain, Buttar and Chintameneni3).
Epidemiological studies have indicated that compared with an intake of SFA, n-6 PUFA intake is associated with a lower risk of CVD(Reference Wang4). Among the latter, linoleic acid (LA) (C18:2, n-6) is considered a potent dietary fatty acid that reduces plasma total cholesterol levels in humans(Reference Bowen, Kris-Etherton and West5). Research has also indicated that diets extremely high in LA (16 % to 29 % of dietary fat) contribute to a significant lowering of plasma low-density lipoprotein cholesterol levels in humans by between 16 % and 22 % compared with diets high in SFA (19 % to 30 % of dietary fat) and low in LA (only 2·3 % to 4·5 % of dietary fat)(Reference Bowen, Kris-Etherton and West5). In animal studies, the structured fat containing LA has been found to show a protective effect against atherosclerosis in C57BL/6J mice when fed high LA diets for 15 weeks(Reference George, Mulkins and Casey6). In contrast, the findings of several studies have shown that an increased intake of LA leads to elevated levels of inflammation, which may be associated with endothelial dysfunction, given that LA is a precursor of arachidonic acid (AA) and inflammatory eicosanoids(Reference Lee, Meisel and de Meijer7). The conversion of LA to AA in healthy humans is about 1 %(Reference Choque, Catheline and Rioux8). An excess LA intake could promote the conversion to produce a non-negligible amount of AA and sequentially intermediates of the pro-inflammatory cascade(Reference Choque, Catheline and Rioux8). The inflammatory impact of the LA intake and conversion has been observed in experimental studies using the rat as the model(Reference Marchix, Choque and Kouba9). In vivo, the pro-inflammatory marker isoprostane is formed primarily via the non-enzymatic free radical-catalysed oxidation of AA. Turpeinen et al. have shown that in healthy subjects, urinary 8-iso-prostaglandin F2α (8-iso-PGF2α), a reliable biomarker of inflammation, increases with consumption of an LA-rich diet(Reference Turpeinen, Basu and Mutanen10). Collectively, the findings of the aforementioned studies indicate that an increased intake of LA leads to increased inflammation and oxidative stress, which may be associated with endothelial dysfunction and promotes the development of atherosclerosis in animal models.
LA is in ambivalent status on the serum cholesterol-lowering and aortic pro-inflammation. The findings of a previous study by Sato et al. have revealed that LA-rich safflower oil reduces the development atherosclerosis beyond its oxidative and inflammatory stress-increasing effects in apoE-deficient mice(Reference Sato, Shibata and Nomura11). The essence of this dietary LA activity is assumed to be associated with its effect on lowering serum cholesterol levels. Consequently, dietary LA may promote arteriosclerosis when serum cholesterol levels are comparable to those of a SFA diet by controlling the cholesterol content of the LA diet. In the present study, we used apoE-deficient (ApoE-/-) mice fed a high-cholesterol diet as an atherogenic animal model to investigate whether LA can induce the development of aortic atherosclerosis in response to the provision of a high-cholesterol LA diet.
Materials and methods
Materials
Trimethylsilylation reagent solution, a mixture of N, O-bis (trimethylsilyl) trifluoroacetamide and trimethylchlorosilane (99:1), was purchased from Merck-Supelco. Commercial enzyme assay kits (Cholesterol E, Free Cholesterol E and Triglyceride E test kits) were purchased from Fujifilm Wako Pure Chemical Corporation. 15-F2t-IsoPM [18O2], as a stable isotope labelled 15-F2t-IsoPM which is used for analysing the content of 15-F2t-IsoPM in urine by GC-MS, was kindly donated by Dr J.D. Morrow (Department of Medicine and Pharmacology, Vanderbilt University, Nashville, USA).
Animals and diets
ApoE-/- mice were purchased from the Jackson Laboratory. A total of 30 male and female ApoE-/- mice (18 males and 12 females, 7 weeks old) with initial weights of 20·5 and 19·1 g, respectively, were raised in the Animal Laboratory Building of the Department of Bioresources and Bioenvironmental Science, Faculty of Agriculture, Kyushu University. The mice were individually housed in metal cages and maintained in a temperature-controlled room (22–25°C) under a 12-h light cycle (dark period 20.00–08.00). Prior to the experimental period, the mice were given free access to a commercial non-purified diet (NMF; Oriental Yeast Co.) and deionised water for 1 week. Following this preliminary phase, the thirty male and female ApoE-/- mice were randomly divided into three groups (each containing six males and four females). These mice were fed one of the following three AIN-76 diets containing different fat and cholesterol contents: SFA-rich fat (50 g palm oil and 50 g lard/kg) and low cholesterol content (0·4 g/kg) (the SFA group); LA fat (100 g high-linoleic safflower seed oil/kg) and low cholesterol content (0·4 g/kg) (the LALC group); and LA fat (100 g high-linoleic safflower seed oil/kg) and high cholesterol content (1 g/kg) (the LAHC group). The overall and fatty acids compositions of the three experimental diets are listed in Tables 1 and 2, respectively. In the feeding experiment, the male and female mice were fed an experimental diet for 9 weeks, after which the animals were then sacrificed by withdrawal of aortic blood under pentobarbitone anesthesia (0·08 mg/100 g body weight), followed by immediate excision of the liver and the aorta for use in subsequent experiments. To isolate serum, blood was collected, incubated at room temperature for 30 min and centrifuged at 1750 × g for 10 min at 4°C. The handling and euthanasia of all animals were carried out in accordance with nationally prescribed guidelines, and all experiments were conducted following ethical approval from the Animal Care and Use Committee of Kyushu University (authorisation number: A02–005–01), as well as in accordance with law no. 105 and notification no. 6 of the Government of Japan.
Table 1. Compositions of the experimental diets
LALC, LA fat and low cholesterol content; LAHC, LA fat and high cholesterol content.
Table 2. Fatty acid compositions of the experimental diets
LALC, LA fat and low cholesterol content; LAHC, LA fat and high cholesterol content.
Morphometric determination of atherosclerosis
ApoE-/- mice were perfused with 50 ml of phosphate-buffered saline (pH 7·4) via a cannula inserted into the left ventricle, and following dissection of the aorta and its main branches from the aortic valve to the iliac bifurcation, perfusion of the heart was immediately continued with 50 mL of 10 % (v/v) neutral formalin buffer solution (pH 7·4). The heart was removed and fixed in 10 % (v/v) neutral formalin buffered solution and stored at 4°C for approximately 1 week. To determine the size and cross-sectional area of aortic lesions, hearts containing aortic roots were processed for quantitative atherosclerosis assays as previously described by Tomoyori et al. (Reference Tomoyori, Kawata and Higuchi12)
Analysis of serum and liver parameters
The levels of total cholesterol in serum were measured using commercial enzyme assay kits (Cholesterol E test kit), and to determine HDL cholesterol levels, HDL fractions were collected as described previously by Tomoyori et al. (Reference Tomoyori, Kawata and Higuchi12) and then measured using the Cholesterol E test kit. Total lipids from the liver were extracted using the method described by Folch et al. (Reference Folch, Lees and Stanley13), and total liver and free cholesterol levels were measured using a method previously described by Tomoyori et al. (Reference Tomoyori, Kawata and Higuchi12). Liver cholesterol ester levels were calculated as the difference between total and free cholesterol levels. Liver triacylglycerol levels were measured using a Triglyceride E test kit, liver phospholipids were chemically determined as described previously by Tomoyori et al. (Reference Tomoyori, Kawata and Higuchi12) and the fatty acid composition of hepatic phospholipids was determined according to the methods described by Carvajal et al. (Reference Carvajal, Nakayama and Kishi14).
Determination of aortic Monocyte Chemoattractant Protein-1 (MCP-1) mRNA levels
Total RNA was isolated from the aorta using the phenol/chloroform method, as described previously by Yuan et al. (Reference Yuan, Minobe and Tanaka15). Complementary DNA was synthesised from 1·0 μg of total RNA using a Transcriptor First Strand cDNA Synthesis Kit (Roche). Gene expression levels were analysed based on quantitative real-time reverse transcription PCR using a SYBR® Premix EX Taq II Kit (Takara) and an Mx3000P QPCR Thermal Cycler System (Agilent). mRNA levels were normalised using the β-actin (Actb) gene as an internal standard. The sequences of primers used for analyses are showed as follows: MCP-1 (Forward primer: CACCAGCAAGATGATCCCAATG and Reverse primer: AAGGCATCACAGTCCCAGTCACAC); and Actb (Forward primer GGATCCCAGATCATGTTTGAGACCTTCAA and Reverse primer: GAATTCGGAGAGCATAGCCCTCGTAGATGG).
Determination of urinary isoprostaglandins
Purification and measurement of urinary 8-iso-PGF2a and 2,3-dinor-5,6-dihydro-8-iso-prostaglandin F2a (dinor) were carried out using a Shimadzu QP2020A GC/MS system (Shimadzu) as previously described by Sato et al. (Reference Sato, Shibata and Nomura11).
Statistical analysis
All data are presented as the means and standard error of the mean (sem) unless otherwise indicated. Male and female groups were compared using one-way ANOVA with Fisher’s protected least significant difference test (PLSD) test to identify individual differences. Analyses were performed using GraphPad 8.3 with significance set at P < 0·05, and 0·05 < P < 0·1 being considered indicative of a trend. In subsequent analyses, we did not distinguish between the sexes, and analysis of the correlation between related indicators was performed using a correlation set at P < 0·05.
Results
Growth parameters and relative organ weights
As shown in Table 3, there were no significant differences in dietary intake, feed efficiency, liver weight or aorta weight between males and females among the three groups. In male mice, the final body weight was significantly higher in those in the LALC group than in the LAHC group, and body weight gain tended to decrease in the LAHC group (P = 0·052) compared with that in the LALC group. Contrastingly, there are no significant differences in body weight gain or final body weight among female mice in the three experimental groups.
Table 3. Growth parameters assessed in the study (Mean values with their standard errors of the mean, n 4 or 6)
* b.w., body weight; LALC, LA fat and low cholesterol content; LAHC, LA fat and high cholesterol content.
Different superscript letters indicate a significant difference at P < 0·05, as determined using Fisher’s PSLD method.
Lesion size and cross sections in the aortic root
As shown in Fig. 1, in male mice, the size and cross sections of aortic root lesions were significantly reduced in the LALC group compared with the SFA group and were significantly increased in the LAHC group compared with the LALC group. However, in female mice, there were no significant differences among females with respect to lesion size and cross section between SFA and LALC groups, although these parameters were significantly increased in the LAHC females compared with those in the LALC group.
Fig. 1. Lesion sizes in the aortic root (a) and aortic root cross sections (b) from male and female ApoE-/- mice. Values are the means ±sem (n 4 or 6). Different letters above bars indicate a significant difference at P < 0·05, as determined using Fisher’s PSLD method. LALC: LA fat and low cholesterol content; LAHC: LA fat and high cholesterol content.
Serum, hepatic and urinary parameters
As shown in Table 4, in male mice, serum and hepatic total cholesterol levels were significantly lower in the LALC group than in the SFA group and significantly higher in the LAHC group than in the LALC group. Hepatic cholesterol ester levels were significantly lower in the LAHC group than in the LALC group. Analogous results were obtained for the female groups with respect to serum and hepatic total cholesterol levels. Urinary 8-iso-PGF2α levels in the female LALC group were found to be significantly higher than those in the LAHC group. There was no significant difference among the three male groups with respect to urinary 8-iso-PGF2α levels. Conversely, in male mice, urinary dinor levels were significantly higher in the LALC group than those in the LAHC group, whereas no significant difference in urinary dinor levels were detected among the three female groups.
Table 4. Serum cholesterol, liver lipid and urinary isoprostane levels (Mean values with their standard errors of the mean, n 4 or 6)
8-iso-PGF2α, * 8-iso-prostaglandin F2α; dinor, **2,3-dinor-5,6-dihydro-8-iso-prostaglandin F2α. LALC, LA fat and low cholesterol content; LAHC, LA fat and high cholesterol content.
Different superscript letters indicate a significant difference at P < 0·05, as determined using Fisher’s PSLD method.
Aortic MCP-1 mRNA level
As shown in Fig. 2, in male mice, the levels of aortic MCP-1 mRNA were significantly higher in the LALC group than in the SFA group, whereas no significant differences between the male LALC and LAHC groups were observed. In female mice, consumption of the different experimental diets had no effects on MCP-1 mRNA levels.
Fig. 2. Aortic MCP-1 mRNA levels. Values are the means ± sem (n 4 or 5). Different letters above bars indicate a significant difference at P < 0·05, as determined using Fisher’s PSLD method. LALC: LA fat and low cholesterol content; LAHC: LA fat and high cholesterol content.
Fatty acid composition of liver phospholipids
As shown in Table 5, both in male and female mice, the levels of AA (C20:4 n-6) were significantly higher in the LALC and LAHC groups than in the SFA group, and a similar pattern was observed regarding n-6 PUFA. We also found that the 20:4/18:2 ratio in the male LAHC group was significantly lower than that in the LALC group, whereas no significant difference in this ratio was detected between the female LALC and LAHC group.
Table 5. Fatty acid composition of hepatic phospholipids (Mean values with their standard errors of the mean, n 4 or 6)
LALC, LA fat and low cholesterol content; LAHC, LA fat and high cholesterol content.
Different superscript letters indicate a significant difference at P < 0·05, as determined using Fisher’s PSLD method.
Discussion
In this study, we established that compared with SFA, dietary LA intake was associated with a significant prevention of atherosclerosis development in male and female ApoE-/- mice only in the condition of the lower content (0·04 % in the diet) of dietary cholesterol. However, this positive effect of LA on atherosclerosis development was found to be negated by consumption of a diet containing a high content of cholesterol. Imaizumi reviewed the relationship between dietary components and atherosclerosis in ApoE-/- mice(Reference Imaizumi16). The analysis of fifty studies were examined that the relationship between serum cholesterol levels and aortic lesion sizes in ApoE-/- mice fed fifty different kinds of dietary components and a correlation coefficient (r) of 0·43 (P < 0·01) was obtained for the association between serum cholesterol levels and aortic lesion sizes(Reference Imaizumi16). Notably, compared with other effective dietary supplements, supplementation with LA was found to contribute to the lowest level of serum cholesterol, although gave rise to the largest aortic lesions. We consider that, as a precursor of AA and inflammatory eicosanoids, an increase of dietary LA content leads to an increase in aortic inflammation and further promotes the development of atherosclerosis. Saini and Keum have reported that the n-3 and n-6 PUFA ratio in a diet strongly affects eicosanoid production, and a reduced consumption of n-3 fatty acids contributes to an enhanced synthesis of inflammatory eicosanoids from AA(Reference Saini and Keum17). In the present study, the fatty acid contents of the LALC and LAHC experimental diets comprised 76% n-6 PUFA, which would further promote an inflammatory response. Accordingly, serum cholesterol levels and aortic root inflammation are considered to be the most important pathogenic factors contributing to the development of atherosclerosis in ApoE-/- mice.
In the present study, we observed that differences in the levels of serum cholesterol are associated with differences in the size of aortic lesions in both male and female ApoE-/- mice. Specifically, we detected significant increases in serum cholesterol levels and aortic root lesion sizes in male and female mice fed the LAHC diet compared with those fed the LALC diet (Fig. 1, Table 4). Such elevated serum cholesterol levels are considered a high-risk pathogenic factor for atherosclerosis(Reference Douglas and Channon18). Animal model studies have shown that a high-dietary cholesterol intake increases total serum cholesterol levels in ApoE-/- mice and ExHC rats(Reference Duivenvoorden, Voshol and Rensen19,Reference Tanaka, Ono and Miyago20) . Additionally, compositions of dietary fatty acids are also an important factor for serum cholesterol levels. Generally, dietary fatty acids can both increase and decrease when dietary oils are changed. In this study, the change in oils, from palm oil and lard to high-LA safflower oil, increased LA in the diet, but at the same time it was accompanied by a decrease in SFA and oleic acid. Therefore, it is quite difficult to conclude the cause of the change in serum cholesterol levels to the change in the composition of a single fatty acid. Yu et al. established a multiple regression formula by meta-analysis of epidemiological studies on the effects of changes in dietary fatty acid composition on serum cholesterol levels(Reference Yu, Derr and Etherton21). It is not suitable to analyse the results of animal experiments with the equation established from human epidemiological studies. However, it is considered that the weighting of each fatty acids does not change significantly between species. Based on that, the effects of dietary fatty acid changes were estimated. In our research, the major changes in dietary concentrations of SFA (mainly palmitic acid (C16:0) and stearic acid (C18:0)) and MUFA (mainly oleic acid (C18:1)) caused the changes of serum cholesterol levels between SFA and LALC groups. However, considering the coefficient of ⊿PUFA in the formula of Yu is negative. LA levels were increased sevenfold in the LALC group compared with the SFA group. We conclude that the levels of LA increase also contributed to serum cholesterol decreased between SFA and LALC groups. At the same time, Keys et al. also reported that changes in serum cholesterol levels are affected by changes of dietary SUFA, PUFA and cholesterol levels(Reference Keys and Parlin22). The dietary SFA and PUFA levels are the same between LALC and LAHC groups. The serum cholesterol levels increased only due to the increase of dietary cholesterol between LALC and LAHC groups. We demonstrated that dietary cholesterol contributed directly to an increase in serum cholesterol levels and exacerbated atherosclerosis in mice. Moreover, aortic lesion size was found to be positively correlated with serum cholesterol levels (Supplemental Table). Cooper et al. found that conjugated LA had no effect on atherosclerosis but had adverse effects on lipoprotein and liver lipid metabolism in ApoE-/- mice fed a 1·25 % cholesterol diet(Reference Cooper, Miller and Mitchell23). We confirmed that dietary high-cholesterol intake had the effect of negating the serum cholesterol level-lowering effect of LA, thereby exacerbating the development of atherosclerosis.
A large number of studies have reported the increase of inflammatory biomarkers such as C-reactive protein (CRP), Tumor necrosis factor (TNF) and Inducible nitric oxide synthase (iNOS) as the development of atherosclerosis(Reference Tuttolomondo, Di Raimondo and Pecoraro24). However, in the present study, we selected aortic MCP-1 mRNA as a marker of aortic inflammatory response. MCP-1, the aortic levels of which are considered an index of the inflammatory response of the vascular wall, is transcriptionally regulated by NF-kB, a transcriptional regulator that responds to pro-inflammation(Reference Panee25). Therefore, aortic MCP-1 mRNA was considered to represent the inflammatory status in the vascular wall more accurately than other serum markers. Arterial inflammation is associated with atherosclerosis, and this association has been found to be sex-dependent(Reference Fiechter, Haider and Bengs26). In male mice, we found that male mice fed the LALC diet were characterised by elevated levels of aortic MCP-1 mRNA, thereby indicating an aortic inflammatory response, whereas in contrast, no similar response was detected in male mice provisioned with the LAHC diet (Fig. 2). We confirmed that the higher levels of cholesterol in the LAHC diet appeared to have no significant effect on aortic inflammation in male ApoE-/- mice. It has been reported that endogenous AA or inflammatory eicosanoids are transported from the liver to the aorta by lipoproteins and albumin, thereby promoting the development of atherosclerosis(Reference Stein and Stein27). Our observations indicated that LAHC diet not promoting inflammatory responses in the aortic root may conceivably be attributed to a reduction in liver-derived AA, and consequently, we further analysed the composition of hepatic fatty acids. The AA in hepatic phospholipids is derived from inflammatory eicosanoids and is naturally incorporated into structural phospholipids in the cell membrane(Reference Hanna and Hafez28). In the present study, we analysed the fatty acid composition of hepatic phospholipids as a means of evaluating AA synthesis as a desaturation index, and accordingly found that consumption of the LAHC diet had the effect of significantly reducing the levels of hepatic AA and urinary dinor (an AA metabolite), and contributed to a reduction in the hepatic 20:4/18:2 ratio. LA is acted upon by desaturation enzymes (delta-6 fatty acid desaturase and delta-5 fatty acid desaturase) and is subsequently converted to AA in the liver(Reference Hastings, Agaba and Tocher29). Several studies have reported that high-cholesterol intake suppresses the action of these desaturases during the conversion of LA to AA(Reference Das30). We thus assume that the high cholesterol levels in LAHC diet inhibit the conversion of LA to AA in the livers of male mice. High level of cholesterol in this diet did not aggravate the aortic inflammatory response due to inhibition of AA synthesis in the liver, which further reduces AA transported to the aorta. In contrast to the male mice, we found that the different experimental diets had no significant effect on aortic MCP-1 mRNA levels in female mice. Moreover, the arterial MCP-1 mRNA levels in male mice were found to be significantly higher than those in female mice. These phenomena indicated that arterial inflammation in female mice was reduced compared with that in male mice. We believe that this reduction in arterial inflammation in females might be attributed to the action of oestrogen, which has been established to be a regulator of inflammation and exerts anti-inflammatory effects(Reference Nilsson31). There are a few reports about using female animal models to evaluate atherosclerosis. 17β-estradiol treatment has been shown to suppress MCP-1 protein and gene expression in ovariectomised (OVX) rabbits(Reference Pervin, Singh and Rosenfeld32), and oestrogen treatment has been found to attenuate vascular re-modelling by promoting the synthesis of MCP-1 mRNA in endothelial cells(Reference Nakagami, Nakagami and Osako33). Moreover, it has been demonstrated that oestrogen reduces MCP-1 mRNA levels in female OVX apoE-/- mice, thereby preventing the development of atherosclerosis(Reference Sato, Sato and Ogawa34). On the basis of the foregoing discussion, it is thus reasonable to assume that high-dietary cholesterol of LAHC diets does not aggravate the aortic inflammatory response in either male or female ApoE-/- mice, although there may be differences in the underlying mechanisms. Moreover, we established that the size of aortic lesions size shows no appreciable correlation with the levels of either aortic MCP-1 mRNA or urinary isoprostane (Supplemental Table). We accordingly speculate that the increase in aortic lesion size and cross section are attributable to elevated levels of serum cholesterol, but not aortic inflammation. Consequently, serum cholesterol levels might be considered a more pathogenic factor than aortic inflammation in promoting atherosclerosis development.
At present, LA is still controversial in terms of cardiovascular prevention and pro-inflammation in humans. GH Johnson and K Fritsche have reported that no evidence was available from randomised, controlled intervention studies among healthy, non-infant human beings to show that addition of LA to the diet increases the concentration of inflammatory markers(Reference Johnson and Fritsche35). Unfortunately, no clinical study in which LA is the unique variable is available, which would allow to conclude on its pro-inflammatory effects. There is also no research reported that high-dietary LA intake increases the concentration of inflammatory markers among cardiovascular patients. On the other hands, anti-effects of CVD by dietary LA were discovered with observation studies(Reference Ramkumar and Sharma36). However, it has not been designed that dietary LA directly affects inflammatory status in human nutritional intervention studies. We expect that our study provide a research direction or theoretical basis for future clarification of the physiological role of LA in humans.
In conclusion, the findings of this study indicate that dietary intake of high cholesterol negates the serum cholesterol concentration-lowering effect of LA and has no significant effect on the inflammatory response on aortic root vessel walls in either male or female ApoE-/- mice. We believe that the inhibitory effect of LA on arteriosclerosis is cancelled by higher dietary cholesterol due to a direct increase in the level of serum cholesterol.
Acknowledgements
X.Y was supported by JST SPRING (grant number: JPMJSP2136). The authors also would like to thank Editage (www.editage.jp) for English language editing.
This research was supported in part by a Grant-in-Aid for Scientific Research (B) (Masao Sato, grant number: 11460060).
Xingyu Yuan: Investigation; Data curation; Formal analysis; Validation; Visualization; Writing—original draft; Writing—review& editing. Rika Nagamine: Data curation ; Investigation; Validation; Formal analysis. Yasutake Tanaka: Investigation; Methodology; Validation; Writing—review& editing. Wei-Ting Tsai: Writing—review& editing ; Validation. Zhe Jiang: Formal analysis; Validation. Ai Takeyama: Validation. Katsumi Imaizumi: Conceptualization; Methodology; Project administration; Resources. Masao Sato: Conceptualization; Funding acquisition; Methodology; Project administration; Resources; Supervision; Writing—review& editing
The authors declare that they have no conflicts of interest.
